1. Field of the Invention
This invention relates to an image reader for reading an original image by an image sensor and the like, particularly to an image reader installing a CIS (Contact Image Sensor).
2. Description of the Related Art
Hitherto, an image reader for automatically reading image information on both sides of an original without the intervention of the user (automatic double-side reader) has been widely used as a reader of a copier, a facsimile, etc., or a scanner for computer input. As the automatic double-side readers, a method of reversing the side of an original in an original reversal section and reading image information is most widely adopted. That is, to input image information on both sides in the method in the related art, one side is read in an original read section and then the side of the ejected original is reversed and again is transported to the original read section for reading the other side.
However, in the automatic double-side reading by inverting the side, it is necessary to reverse the original after once being ejected and again transport the original to the original read section and thus it takes much time in reading both sides and the productivity is poor in reading both sides. The original reversal section requires a complicated mechanism to reverse the side of an original and the original jam occurrence percentage in the original reversal section is high as compared with other transport sections. Improving the reliability is required. Further, to design an automatic double-side reader that is placed in a narrow space, the necessity for inverting an original rapidly in a small diameter may occur because of the necessity for inverting the original and collating the pages of the original at the paper ejection time, etc. Consequently, it is difficult to transport an original of a predetermined cardboard having a large basis weight.
An art of automatically reading both sides in one transport is studied. For example, in JP-A-1-171360 and JP-A-1-293757, two image sensors are placed on both sides of an original path for transporting an original and both sides of an original can be read automatically in one original transport without reversing the side of the original.
Generally, to read an original, for example, a technique of applying light from a light source of a fluorescent lamp to the original and reading the reflected light by an optical sensor through a reduction optical system is adopted. As the sensor in the technique, for example, a one-dimensional CCD (Charge Coupled Device) sensor is used to process one line at a time. Upon completion of reading one line in the line direction (main scanning direction), the original is moved a minute distance in the direction orthogonal to the main scanning direction (subscanning direction) and the next line is read. This operation is repeated over the whole original size for completing reading of one page of the original. As a method of reading an original in order in the subscanning direction without moving the original, a method of moving a plurality of mirrors by a moving body such as a full-rate carriage or a half-rate carriage for reading an original in order in the subscanning direction is also available.
In the read technique, it is necessary to apply a light source to an original and reading the reflected light by a CCD sensor through several mirrors as described above and thus the whole unit tends to become large. Particularly, if a plurality of image sensors need to be placed to read both sides without reversing an original, it is difficult to place a plurality of CCD sensors because of a space problem. To solve the space problem, use of an image sensor called CIS (Contact Image Sensor) for directly reading an image by a linear sensor through a SELFOC lens using a small LED (Light Emitting Diode) as a light source is studied.
However, in the read technique using the CIS, the focal depth is very shallow and if the CIS and an original face go out of distance even a little, out-of-focus occurs and the read image becomes unclear. That is, it becomes hard to provide a sharp image although the whole unit can be reduced.
To lessen defocus as much as possible, conventionally, an original is pressed against a glass face disposed on CIS and is transported and the original pressed against the glass face is focused. However, if the original is pressed against the glass face and is transported, pencil carbon, ink, garbage, dirt, etc., are pressed against the glass face and rub the glass face, and are is easily deposited on the glass face. If the read image is output with the pencil carbon, ink, garbage, dirt, etc., deposited on the glass face, a black line in the original transport direction (subscanning direction) appears and measures against image quality trouble must be taken.
Further, to adopt both a scanner for applying a light source to an original and reading reflected light by a CCD sensor through several mirrors and a reader using a CIS and make it possible to read both sides of the original at the same time, the reader using a CIS is generally placed to direct downward and the glass thereof is placed on the upper side relative to an original transport passage. In such a case, the user cannot sufficiently clean the dirty glass surface. Particularly, if the glass face in the reader using a CIS cannot be opened because of the structure, it becomes difficult for the user to clean the glass surface.